[GH-ISSUE #23746] issue: Data truncated when uploaded as a file, no issues when the file text is pasted into the prompt #90795

Closed
opened 2026-05-15 16:04:24 -05:00 by GiteaMirror · 3 comments
Owner

Originally created by @huornlmj on GitHub (Apr 15, 2026).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/23746

Check Existing Issues

  • I have searched for any existing and/or related issues.
  • I have searched for any existing and/or related discussions.
  • I have also searched in the CLOSED issues AND CLOSED discussions and found no related items (your issue might already be addressed on the development branch!).
  • I am using the latest version of Open WebUI.

Installation Method

Docker

Open WebUI Version

v0.7.2 (latest)

Ollama Version (if applicable)

N/A

Operating System

Linux

Browser (if applicable)

No response

Confirmation

  • I have read and followed all instructions in README.md.
  • I am using the latest version of both Open WebUI and Ollama.
  • I have included the browser console logs.
  • I have included the Docker container logs.
  • I have provided every relevant configuration, setting, and environment variable used in my setup.
  • I have clearly listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc).
  • I have documented step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation. My steps:
  • Start with the initial platform/version/OS and dependencies used,
  • Specify exact install/launch/configure commands,
  • List URLs visited, user input (incl. example values/emails/passwords if needed),
  • Describe all options and toggles enabled or changed,
  • Include any files or environmental changes,
  • Identify the expected and actual result at each stage,
  • Ensure any reasonably skilled user can follow and hit the same issue.

Expected Behavior

When an HTML file is uploaded via open WebUI, the models should be able to observe all of the content in it without it being truncated.

Actual Behavior

The models are not able to see content in the file roughly past the mid point. If I just copy the full text from the HTML file and past it in with the prompt as normal, the model is able to see and cite all parts of the text, so it's not a context window issue. The file upload must be truncating the document.

Steps to Reproduce

  1. May not be reproducable as I cannot share the actual content of the html file that I am seeing the issue in. But you can try a standard markdown - plain text webpage without images or javascript.
  2. If the document has bullet points and has around 20 sentences per paragraph that might work.
  3. Make the document have around 7 numbered headings.
  4. Upload the document as a file and ask the models what point 7 is about.
  5. It will say something like "Based on the provided text, there is no heading 7. The numbered sections in the document only go up to "4."

Logs & Screenshots

Image

The original html document has 7 numbered sections.

Additional Information

No response

Originally created by @huornlmj on GitHub (Apr 15, 2026). Original GitHub issue: https://github.com/open-webui/open-webui/issues/23746 ### Check Existing Issues - [x] I have searched for any existing and/or related issues. - [x] I have searched for any existing and/or related discussions. - [x] I have also searched in the CLOSED issues AND CLOSED discussions and found no related items (your issue might already be addressed on the development branch!). - [x] I am using the latest version of Open WebUI. ### Installation Method Docker ### Open WebUI Version v0.7.2 (latest) ### Ollama Version (if applicable) N/A ### Operating System Linux ### Browser (if applicable) _No response_ ### Confirmation - [x] I have read and followed all instructions in `README.md`. - [x] I am using the latest version of **both** Open WebUI and Ollama. - [x] I have included the browser console logs. - [x] I have included the Docker container logs. - [x] I have **provided every relevant configuration, setting, and environment variable used in my setup.** - [x] I have clearly **listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup** (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc). - [x] I have documented **step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation**. My steps: - Start with the initial platform/version/OS and dependencies used, - Specify exact install/launch/configure commands, - List URLs visited, user input (incl. example values/emails/passwords if needed), - Describe all options and toggles enabled or changed, - Include any files or environmental changes, - Identify the expected and actual result at each stage, - Ensure any reasonably skilled user can follow and hit the same issue. ### Expected Behavior When an HTML file is uploaded via open WebUI, the models should be able to observe all of the content in it without it being truncated. ### Actual Behavior The models are not able to see content in the file roughly past the mid point. If I just copy the full text from the HTML file and past it in with the prompt as normal, the model is able to see and cite all parts of the text, so it's not a context window issue. The file upload must be truncating the document. ### Steps to Reproduce 1. May not be reproducable as I cannot share the actual content of the html file that I am seeing the issue in. But you can try a standard markdown - plain text webpage without images or javascript. 2. If the document has bullet points and has around 20 sentences per paragraph that might work. 3. Make the document have around 7 numbered headings. 4. Upload the document as a file and ask the models what point 7 is about. 5. It will say something like "Based on the provided text, there is no heading 7. The numbered sections in the document only go up to "4." ### Logs & Screenshots <img width="1398" height="282" alt="Image" src="https://github.com/user-attachments/assets/ebc5e576-70a6-4345-9ec8-a9b153fad20b" /> The original html document has 7 numbered sections. ### Additional Information _No response_
GiteaMirror added the bug label 2026-05-15 16:04:24 -05:00
Author
Owner

@Classic298 commented on GitHub (Apr 15, 2026):

0.7.2 is not the latest and this looks like a user error. did you enable full context mode? looks like the RAG was at work

<!-- gh-comment-id:4253227359 --> @Classic298 commented on GitHub (Apr 15, 2026): 0.7.2 is not the latest and this looks like a user error. did you enable full context mode? looks like the RAG was at work
Author
Owner

@huornlmj commented on GitHub (Apr 15, 2026):

@Classic298, thanks. I enabled full context mode. But there is an issue in the version tool. It clearly shows to me that 0.7.2 is the latest every time I click "Check for updates"

Image
<!-- gh-comment-id:4253885274 --> @huornlmj commented on GitHub (Apr 15, 2026): @Classic298, thanks. I enabled full context mode. But there is an issue in the version tool. It clearly shows to me that 0.7.2 is the latest every time I click "Check for updates" <img width="279" height="217" alt="Image" src="https://github.com/user-attachments/assets/cf2df6ad-22c1-4d8d-997e-192bcd1f5d04" />
Author
Owner

@Classic298 commented on GitHub (Apr 15, 2026):

You enabled offline more or disabled version Update checks.

You are on Open WebUI GitHub - go to the releases section. You will see you are 13 versions behind

<!-- gh-comment-id:4253945990 --> @Classic298 commented on GitHub (Apr 15, 2026): You enabled offline more or disabled version Update checks. You are on Open WebUI GitHub - go to the releases section. You will see you are 13 versions behind
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/open-webui#90795